Antonius Block, thanks for a very informative post.
You are certainly very knowledgeable about these matters, but if you say that building HSR on the East Coast would price most people out of using it, are you saying that there is some reason that ticket prices would be cheaper in California?
I read his arguments as the price increase on the East Coast would be much higher to the relative benefit in time savings. A rail link between NYC and DC, per his cites, already runs either 2:48 for Acela, or 3:12 for regular rail. And the price differential for that 24 minutes of saved time is almost $100. The cost is, according to him (and in my opinion, too), already too high for the marginal benefit it accrues.
On the west coast, the non-air links between the cities mentioned are cheaper, but also much longer. The two time frames mentioned are almost 9 hours or almost 13 hours. Doubling, or trebling, the ticket cost while reducing transit time by as little as a third, would offer a much greater return for the increased ticket price, I’d think.
As for thinking that the west coast ticket costs remaining substantially lower per track mile - I think it’s fair to make that argument, considering that it’s already a reflection of the disparity between the two rail links. SF to LA (per Google, so take it with grain of salt) is 382 miles. NYC to DC is 229 miles. So, back of the envelope calculations here: the prices quoted in Antonius Block’s post reflect a cost of about $0.14 per track mile; and the NYC to DC link reflects a cost of about $0.43 per track mile. I can’t say, exactly why there’s such a difference in cost - I’d guess some of the difference is the population density on the routes and the number of stations on the routes, but those are just guesses.
Given those figures, however, it doesn’t seem unreasonable for someone to suggest that the pricing will continue to follow that established pattern.
If it included freight it would be more logical. Trucking is one of the reasons we built so many roads. It would save a lot of time and eliminate a lot of danger if we could send freight across country rapidly and safely. It would certainly increase the revenue stream.
gonzomax, mixing freight and HSR passenger can be detrimental to the latter; here is a brief paragraph on how freight is handled on France’s TGV.
One of the main differences is that freight trains are generally much heavier than HSR passenger trains. The latter have an extremely high power/weight ratio, so can climb a much steeper grade than a freight train; when faced with a medium-sized hill, it’s generally more efficient for a passenger HSR to simply go over it and regain some of the energy via regenerative braking on the downhill slope, whereas a heavier freight would be better off going around the hill at lower speed on twistier but lower-gradient track.
Although the above link shows that the Channel Tunnel allows mixed paasenger and freight service, this is usually just through the 50km length of the Tunnel itself. At either end, passenger and freight services generally run on separate tracks so that the freight trains don’t delay the passenger trains any more than necessary.
I could see high-value, low-weight freight (such as overnight package delivery) running on HSR lines – this is done in France with the TGV “La Poste” mail trains, but could also presumably be done with package cars attached to passenger HSR trains. However, the heavy loads that make up most of the US rail-hauled freight are IMHO inconsistent with the high speeds necessary to make passenger HSR a viable proposition.
[Ravenman, sorry not to have replied yet to your question in post #122 regarding the different economics of NEC vs CA-HSR in the US – I’ll try to post something later this evening.]
That’s a very good question, Ravenman, and I hope that I’m up to answering it.
OtakuLoki already covered in post #123 some of what I would have said had I been able to respond earlier.
Following on from my post #119, I’d say that the economic differences between potential HSR systems on the NEC and in California are not only geographical and political, but also historical (i.e. dependent upon past decisions and existing infrastructure):
[ul]
[li]Geographical: IMHO, true HSR needs at least one gap of 100 miles between stops in order for the high-speed trains to “stretch their legs” and get up to the maximum speed. With high-speed tracks, it can actually turn out to be cheaper for the railroad to run high-speed service since many of the costs (crew wages, etc) are proportional to operating time but the revenue is generally roughly proportional to the number of passenger-miles. The density of cities on the NEC makes it hard to find places where HSR trains could really get up to full speed, whereas in California there can be long distances between stations. So, although some trains would stop at Fresno, Bakersfield, etc, it’s quite possible that many trains would be SF - San Jose - LA (with perhaps an additional suburban-LA stop such as Burbank). This would make the average train much faster than geographically feasible on the NEC, which correlates to higher revenue per train (and train crew) per day.[/li]
Outside of the urban areas of the SF Bay Area and the LA basin, land in California is much cheaper than on the NEC – which is why the Central Valley is experiencing so much sprawl at the moment. HSR should encourage higher-density development close to the centers of cities such as Fresno and Bakersfield, which should reduce sprawl and improve the urban environment.
[li]Political: As I said in post #119, choice of routes and specific station stops for NEC-HSR would be a matter of much infighting between the involved states, which leads to suboptimal stopping patterns. CA-HSR will exist within only one state, and therefore should be able to be planned more efficiently.[/li]
[li]Historical (existing infrastructure): The NEC already exists, and is considered the flagship service of Amtrak, an underfunded quasi-federal chimera that has never been noted for its cutting-edge ideas. The current fare structure involves – for a trip from Washington to NYC – a ~$100 price differential between similarly-timed Acela Express and regular trains, for a 24-minute time differential. I don’t know if this is the optimum fare strategy for the NEC, but I cannot imagine such a differential working well on the SF-LA axis.[/li]
In addition, the difference between “Acela Express” and regular NEC trains has much more to do with the trains than any upgrade to the tracks. The new trains “tilt” to allow higher speeds on curves, and the older ones don’t. So, regardless of stopping pattern, there may have been a decision by Amtrak that if the standard trains cost a certain fare ($69-$98 from DC to NYC depending on time of day), then the new trains had to pay for themselves as a separate business model, and so the fares needed to be considerably more (around double) to pay off the investment.
With California-HSR, the situation is totally different. Although there is current Amtrak service between the SF Bay Area and the LA basin (rail via the Coast Starlight, and rail-then-bus via the San Joaquins), the total passenger load is negligible compared to that carried by air. Given the huge time differential between existing train (13 hours or 9 hours) and plane (~1 hour in the air), a realist might claim that there is, to all intents and purposes, no current SF-LA rail competition to the plane.
This tabula rasa might seem like a disadvantage compared to the NEC with all of its existing infrastructure, but IMHO it’s actually a blessing. The two existing north-south rail lines in California (down the Pacific Coast via Santa Barbara; and over the Tehachapi Pass) were paid off long ago, so there’s no real requirement that a CA-HSR network would follow existing routes in order to justify recent expenses[sup]1[/sup]. Since the advanced track would permit much higher speeds, and every CA-HSR train would be capable of the same maximum speeds regardless of its stopping pattern (compared to the NEC where only the “Acela Express” trains can tilt), there’s no reason to believe that the fares will be structured the same way as on the NEC. If a given passenger wants to go all the way from SF to LA, it is actually cheaper for the railroad to put him/her on a nonstop train rather than a stopping one, since it frees up flexibility for intermediate trips on the stopping train. Since CA-HSR would be completely separate from Amtrak and appears committed to yield-management techniques for far pricing, there probably will not be a premium for faster journeys, although there will undoubtedly be different classes of service (similar to airplanes – the Business Class passengers don’t fly any faster than Coach, but they get more legroom and different food options).
[/ul]
Case Study:
Although it’s commuter rail rather than long-distance, Caltrain offers a valuable lesson in revenue optimization between “limited-stop” and “local” service in California.
During the Dot-Com Boom in the Bay Area (late 90s-2001 or so), plans were made to upgrade the Caltrain right of way and add passing tracks to allow limited-stop service between San Francisco and San Jose. By the time the “CTX” project was completed and the “Baby Bullet” service started in 2004, the boom was long over and ridership was down. The Baby Bullets (which were faster than regular service trains due to fewer stops, but had the same fare structure) were popular, but Caltrain was falling short of budget projections. Many networks would have cut the number of trains to reduce costs, or introduced a fare differential between the stopping trains and the Baby Bullets. Caltrain, however, went against the grain and restructured the service to add more limited-stop Baby Bullets. Although a small number of passengers saw their service level decrease, the new system turned out to be an improvement for most passengers, and ridership has gone up more than 10%. The Baby Bullets still have the same fare structure as the slower trains, but run with very good passenger loads – and are cheaper for Caltrain to run (less fuel for accelerating after stops, and fewer employee-hours for a given number of runs)!
It might be worth mentioning (since CA-HSR will share trackage with a further-improved and electrified Caltrain service, and presumably many of the same planners will be involved) that the 2002-2004 CTX upgrade came in on-time and under budget…
[sup]1[/sup]The planned CA-HSR would in fact go through the Tehachapi Pass, but would avoid bottlenecks such as the Tehachapi Loop. The route through the Loop is the busiest single-track line in the world, which is why Amtrak’s San Joaquin trains don’t run south of Bakersfield – the freight railroad that owns the tracks (Union Pacific) doesn’t consider it worthwhile allowing passenger trains to take up valuable time slots, at any price that Amtrak is willing to pay.